Virtual Protocol: AI Factory for Games and Metaverse

Forward the Original Title ‘Virtual Protocol ：为游戏和元宇宙而生的 AI 工厂，人人皆可贡献与受益’

The popularity of the AI ​​track is obvious to all, and it has become a narrative consensus throughout the year.

On the other hand, the transaction volume of the Base chain has surged recently, with TVL reaching a new high of US$745 million.

However, this trend cannot last forever. With the popularity of Base chain traffic and reliable AI projects, the combination of the two may create new opportunities.

Current crypto AI projects often focus on high-profile infrastructure and algorithms that are close to the future but far from everyday life, entertainment, and personal involvement. What we lack is a project that allows individuals to participate, has a low threshold, is perceptible, and has a visible and tangible impact on the entertainment lives of retail investors. Therefore, finding projects with differentiated business in AI is the key to discovering gems.

Recently, Virtual Protocol on the Base chain is a typical example that meets the above combination point.

Its token VIRTUAL quickly doubled in price within one day, and the protocol itself is one of the few serious AI projects on Base.

The goal of the project is to create an AI factory for games and the metaverse, where everyone can contribute and benefit, while making games more intelligent and increasing experience and interaction.

The narratives of GameFi and Metaverse have been dormant for a long time, and perhaps new sparks can be created with AI.

More importantly, game projects can use the capabilities provided by Virtual to make their own games better, so it is foreseeable that Virtual Protocol will form a “golden shovel” effect similar to Pyth.

As AI track projects gradually become fungible, will Virtual become the next Bittensor specializing in the gaming and entertainment industries?

In this issue, we use the most popular way to help you quickly understand this AI pearl on the Base chain.

AI grows towards democratization and games become intelligent

To understand whether Virtual Protocol has potential, we first need to see what is lacking in the market and whether it can solve the problem. Since the project is aimed at the broad direction of AI + Game, what are the problems in the current AI and gaming tracks?

Let’s start with AI. One commonly criticized issue is the black box problem: You contribute data to AI but cannot profit from it yourself, you cannot know how the data is used, and you cannot actively contribute the data you want to contribute; the providers of AI models take away most of the discourse and benefits.

As for games, one of the prominent issues with current GameFi or metaverse is the lack of immersive interactive experience, making it less attractive compared to conventional gaming products. The lack of interactive feeling in the virtual world makes you feel unreal, not fun, not immersive enough, and not personalized enough, and as a result, the game, as a time-consuming product, lacks the confidence to capture players’ attention.

From the perspective of industry observers, we have actually begun to see AI being used in the crypto industry to improve experience and efficiency. However, most of this application is concentrated in DeFi and other transaction-oriented aspects. There are very few mature examples of AI applied to the game track.

And this is also the key for Virtual Protocol to find its suitable ecological niche.

Simply put, this project mainly does two things:

Making AI democratized and games intelligent.

The so-called democratization of AI means that you can contribute the data you wish to contribute, share the benefits of this contribution, understand the purpose of data contribution, confirm the ownership of the data, and break the “black box” mentioned above;

This narrative is naturally in line with the egalitarian and grassroots spiritual core of Web3, and is easy to understand.

The intelligence of the game means that you can use the data contributed by different groups through AI training and integrate it into different game mechanisms, so that all game interactions can be transformed from “quasi-social” to “fully social” —— Imagine “My AI friends and I were playing in the game”. This scenario is gradually becoming a reality;

In the traditional game circle, some players have connected GPT4 to NPCs in “The Elder Scrolls”, and the dialogue quality and interactive immersion have been greatly improved. Web3 games can also adopt this method, making the game more interactive and immersive, and serving the entire game/metaverse track.

Another benefit brought by intelligence is composability. The memory and background of the AI ​​character can be used throughout different games and virtual worlds - he in game A will accompany you on your adventure in game B…

The AI ​​character takes an adventure in Zelda and knows where the sword in Mario may be hidden through memory.

Generative AI helps expand personalization, while multimodal output (including text, sound, and visuals) improves immersion. Together, they remove the monotony from virtual interactions, ensuring each moment is more engaging than the last.

The benefits are self-evident and the narrative is reasonable enough. The key is how to do it.

AI factory for games and metaverse

Back to the project itself, how does Virtual Protocol implement a jointly owned, human-curated, plug-and-play game AI?

Putting aside the high technical details, a popular metaphor can summarize the essence of the project — Build a decentralized AI factory to connect supply and demand.

Since it is a factory, it is natural to think of the key issues at these two ends:

For the supply side, what kind of products does Virtual Protocol have? Where to buy from?

For the demand side, what does Virtual Protocol sell? To whom do you sell the goods?

Let’s start with the supply side.

You can simply understand the goods purchased by Virtual as “Parts of virtual AI characters“, sort the parts into the factory and prepare them for assembly.

The parts can be roughly divided into three types:

• Cognitive core:Incorporating the complexity of large language models (LLMs), it excels at crafting interactions that are not only contextually coordinated but also linguistically sound.
• Sound core: From the timbre and tone of the voice to the ambient sounds that accompany the interaction
• Visual core: Covering everything from basic looks to complex animations. This core encapsulates the visual aspect, enabling the AI ​​to be seen and recognized

Seriously speaking, these parts are officially called Modular Core, and they form the key for a virtual character to have AI capabilities.

Where do these parts come from? The answer is “from the masses”.

Anyone can use their talents and participate in contributing Modular Core to improve an AI virtual character, and then apply the character to various applications such as games.

You can be a text data contributor, providing the necessary textual data to shape the AI’s dialogue, backstory, and characteristics to form a virtual personality;

You can be a LLM model contributor, using target text data to refine language models to enhance AI communications, ensuring natural and contextual interactions. This process transforms general artificial intelligence into more nuanced tools that excel in specific professional and subject areas;

You can still be one voice data contributor, providing different sound samples to create virtual voices that capture the full range of emotions and pitches to generate expressive speech;

You can even be one visual contributor, designing virtual appearances, from basic images to detailed animations.

Finally, the above parts are assembled into an AI virtual character. Everyone contributes, everyone owns, and everyone benefits.

Any materials you contribute will appear in the form of NFT, which facilitates later revenue sharing and data rights confirmation, and the file itself is stored on IPFS.

After the incoming goods and assembly are completed, what goods will be sold?

These AI virtual characters can actually be roughly classified into three types:

• Virtual mirror of IP characters: These virtual mirrors are designed to emulate iconic characters from various IPs. Allowing fans to interact with their favorite characters in a very personal and authentic way, enriching their experience with original IP.
• Function-specific virtual program: Customized to perform specialized tasks, similar to an expert system. Whether weaving horror narratives or guiding players through DOTA, these AIs possess domain-specific knowledge that enables them to act as assistants and autonomous agents, providing users with expertise and interactive learning experiences.
• Personal avatar: It not only looks like you but behaves and interacts in a way that is typical of you. These personal virtual devices are designed to be customizable and self-extending, providing users with the opportunity to interact with digital spaces in an inherently personal and unique way.

Who needs these goods? Naturally, there are various Dapps with different AI requirements.

Virtual Protocol first played a game of “self-produced and self-sold”.

For example, the self-developed virtual AI girlfriend game AI WAIFU uses the capabilities of the above-mentioned AI factory to build three different AI girlfriends that can talk to you, react based on different conversation scenarios, improve your favorability, and form a game development model based on dialogue feedback.

For more specific details, please refer to this article Interpretation of AI WAIFU: The first AI-driven virtual girlfriend game on Blast - the token LBP is about to be launched to get an in-depth introduction.

Similarly, the AI ​​factory can also supply other third-party Dapps.

Through simple SDK access, different application developers can take advantage of the multi-modal reasoning capabilities of the protocol, allowing VIRTUAL to be seamlessly integrated into various DApp environments, ensuring developers can implement advanced AI functions with minimal friction.

There is a key benefit here, which is the “plug and play” mentioned earlier.

You make the logic of the game yourself. When you need AI capabilities, you don’t have to develop it from start to finish yourself, nor do you have to train an AI separately for the game. You can just use the capabilities provided by Virtual Protocol as a plug.

Ultimately, Virtual Protocol, a set of jointly owned, jointly created, plug-and-play game AI, formed such a value flow for different participants:

Contributors can contribute data, models and materials required in the core of various modules;

Verifiers use professional knowledge to review and verify the data and models provided by contributors to ensure that the output meets the highest standards;

Dapp developers use the above-mentioned contributions and audited data and models, and the revenue generated by users using Dapps will also be shared with verifiers and developers. As a result, a factory specifically built for game AI is built.

Virtual token: the key to coordinating the interests of all parties

For this platform connecting crowds on the left hand and DApp projects like games on the right hand, the economic incentives of the token will be key to whether the AI factory can operate transparently and orderly.

The overall effect of $Virtual is to adjust incentive measures and increase the value of all stakeholders involved. Specifically, $Virtual will act as a utility token for transactions, governance, and staking. Users can access and utilize various features in the virtual environment and contribute to them. Let’s take a look at the value of $Virtual from the perspective of an ordinary user.

Suppose Alice, who is bullish on Virtual Protocol, holds $Virtual. She can choose to stake $Virtual to maintain the security of the entire system while earning staking rewards. Now, the question arises, where should she stake her tokens for maximum benefit?

Alice needs to choose a virtual character within the Virtual system to stake her tokens, as shown below. You can think of it as a staking pool, where the more tokens there are, the more development resources will be allocated around that character.

In reality, the project has associated AI virtual characters with Virtual tokens. The more tokens staked, the better it reflects the popularity of the character. The more popular the character, the more willing contributors are to contribute data (because the potential distributed rewards are greater), so the development resources around that character will also increase.

Essentially, Alice is facing a mass voting mechanism, using tokens to vote and benefiting from the voting. However, after staking tokens for a particular character, completion of the staking also implies responsibility. Alice also needs to act as a Validator, reviewing the various data information uploaded by contributors for the character, such as language, voice, appearance, etc. If she does not possess this review capability, she can choose a delegation staking method similar to ETH, delegating the staking rights to more professional nodes and thereby sharing the rewards.

Later, when other DApps develop games using Virtual’s technology and AI characters, Alice may receive token airdrops for those games because she holds Virtual.

Beyond Alice’s perspective, the VIRTUAL token effectively connects the value flow throughout the entire protocol:

Firstly, VIRTUAL tokens reward contributors who upload and update content for the virtual character’s Modular Core.

Secondly, VIRTUAL tokens reward validators who verify submitted content.

Finally, the Virtual Protocol official will use a portion of the income from DApps using virtual characters to repurchase $VIRTUAL tokens, thereby increasing the token’s value and stabilizing the ecosystem.

In addition to the role of the token itself, its economic model deserves more attention.

Judging from the project’s documents and public information, 40% of the VIRTUAL tokens are owned by the project’s official DAO. Any use of these 40% tokens requires the proposal to be voted on before it can be executed. This also means that the uncirculated part of the VIRTUAL tokens will not be sold under pressure, thereby diluting the value of the tokens in circulation.

On the other hand, the remaining 60% has been fully circulated in the market, and there is also the repurchase mechanism mentioned above. Games associated with the protocol generate revenue and distribute profits to token holders.

Therefore, overall, VIRTUAL’s selling pressure is controllable and has a certain shovel effect.

In addition, VIRTUAL token’s Airdrop page has been launched, and you can also get different airdrop rewards by holding Virtual. Although the specific information has not been announced yet, it has also increased market expectations in disguise.

Driving force to watch in the near future

Public information shows that Virtual Protocol’s mainnet will be officially launched on the evening of the 22nd, at which time users will be able to contribute relevant data and models on its website. In addition, what other good information is worth noting?

First of all, in terms of the general environment, the Base chain is hot and the AI ​​track continues to grow. The superposition of these two factors is undoubtedly a favorable factor.

The second point that is easily overlooked is that although the protocol is on Base, applications on each chain can use Virtual capabilities. Token-based economic incentives and value flow are on-chain, but AI capabilities are off-chain. Using SDK to develop applications does not select L1 or L2. This also greatly increases the number of potential partners and usage space of the protocol, that is, the “Pyth golden shovel effect” we talked about at the beginning of the article.

Perhaps, all games and metaverse projects will usher in an AI spring, and Virtual can just be the first ray of spring breeze.

At the same time, full-chain empowerment also brings about a potential space for “vote buying.”

Readers familiar with DeFi will be familiar with Curve War, which creates demand for veCRV to guide more attention to its liquidity pool.

Virtual Protocol also has the possibility of a Virtual War:

With limited attention and liquidity, different game projects can invest as much xVirtual as possible in the virtual characters they want to use to attract others to stake for higher returns, thus promoting activity and use of their own AI virtual characters.

In addition to these fundamentals, some ecological cooperation is also worthy of interpretation.

Virtual’s official Twitter recently stated that it has become part of Nvidia’s Inception Program. The program is a free program designed to help startups grow faster through cutting-edge technology, opportunities to connect with venture capitalists, and access to NVIDIA’s latest technology resources.

Being associated with NVIDIA and AI, at least it has the spark to fuel the speculative narrative.

At the same time, Virtual’s AI virtual characters can essentially be seen as AI agents specializing in the gaming industry. And the big brother of AI agents, Autonolas ($OLAS), has also collaborated with Virtual. There is room for imagination in terms of resource contributions and token linkage between the two.

Virtual, Extracted from Reality

In conclusion, the name Virtual Protocol reflects the core of its business:

Decentralizing the collection of materials from reality makes the virtual world smarter.

The name Virtual Protocol is interesting enough, and its corresponding business and narrative are sufficient to stimulate more interest from participants.

The virtual world is not completely independent from reality; on the contrary, it gathers various wisdom and efforts from reality. This wisdom and effort deserve to be rewarded appropriately. While benefiting the entertainment industry, it also allows everyone to gain their own happiness and profits.

Disclaimer：

1. This article is reprinted from [TechFlow], Forward the Original Title‘Virtual Protocol ：为游戏和元宇宙而生的 AI 工厂，人人皆可贡献与受益’.All copyrights belong to the original author [TechFlow]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.